Introduction
The monograph Changing Residence: The Geographic Mobility of Elderly Canadians (Northcott, Reference Northcott1988) reviewed the research that had been done up to the mid-1980s on elder mobility in Canada and in other countries including Australia, France, Great Britain, and the United States. The purpose of this article is to review progress in research over the past 25 years on the geographic mobility of older persons with a particular emphasis on Canadian seniors; second, to assess the current status of research on elder migration in Canada; and third, to identify future challenges and unanswered research questions, and to make suggestions for future research.
Prior to the publication of Changing Residence, the geographic mobility of elderly Canadians had been largely neglected. Certainly, an extensive literature existed on geographic mobility; however, most of that literature focused on economically motivated relocation undertaken by the non-elderly population that moved primarily for labour force reasons. In contrast, seniors received little attention because they made up a relatively small proportion of the population, were less likely to move than the non-elderly population, and were less likely to move for economic reasons related to employment. Nevertheless, with population aging – that is, with the increasing proportion of the population that is aged 65 and older – seniors and their patterns of relocation and geographic concentration began to receive more attention.
It is important to study elder migration for a variety of reasons. First, seniors constitute a growing percentage of the population (Turcotte & Schellenberg, Reference Turcotte and Schellenberg2007). Second, the geographic mobility of seniors (and of non-seniors) tends to result in concentrations of seniors in selected locales with consequences for local economies (Bradley & Longino, Reference Bradley, Longino and Uhlenberg2009; Walters, Reference Walters2002). For example, seniors who relocate to a certain locale bring with them their Old Age Security payments provided by the federal government, their Canada Pension Plan and other pension income, and their spending preferences which differ from the spending patterns of non-seniors (Hodge, Reference Hodge1991; Li & MacLean, Reference Li and MacLean1989; Serow, Reference Serow2003; Tucker, Mullins, Béland, Longino, & Marshall, Reference Tucker, Mullins, Béland, Longino and Marshall1992; Warnes, Reference Warnes and Uhlenberg2009). In addition, concentrations of seniors also have implications for local economies in terms of demands placed on tax dollars. For example, given that health care is primarily a provincial responsibility in Canada, concentrations of seniors have implications for provincial health care spending (Moore & Pacey, Reference Moore and Pacey2004). A third reason for studying elder migration is the implications that this phenomenon has for the exchange of support in families, including the kinds and amounts of support provided by seniors to their children and grandchildren as a consequence of elder migration and, vice versa, the kinds and amounts of support provided to seniors by their children and grandchildren (Firbank & Johnson-Lafleur, Reference Firbank and Johnson-Lafleur2007; Moore & Rosenberg, Reference Moore and Rosenberg1997; Sarma, Hawley, & Basu, Reference Sarma, Hawley and Basu2009). For these reasons, it is important to study the implications of the geographic mobility of seniors as this population continues to increase in Canada over the next several decades.
Among the many aspects of the geographic mobility of elderly Canadians reviewed in the 1988 monograph, three are selected here as particularly noteworthy. First, evidence from the 1961, 1971, 1976, and 1981 censuses indicated that an increasing number of elderly Canadians were moving from one province to another. In 1981, 1.8 per cent of seniors (some 39,000 seniors) had changed their province of residence at some time between 1976 and 1981. Given that interprovincial migration has resulted in concentrations of elderly persons in popular destinations, for example, in parts of British Columbia and Ontario, the research literature will be examined to determine whether this trend has continued into the early twenty-first century.
Second, literature reviewed in the 1988 monograph pointed to the high volume of elderly seasonal migrants (snowbirds) from Canada wintering in U.S. Sunbelt states such as Florida and Arizona. The research literature over the past 25 years will be reviewed to examine patterns of temporary seasonal migration of elderly Canadian snowbirds to the United States and their impact on health and social care systems.
Third, the in-migration of elders from outside of Canada was noted in the 1988 study. Over 30,000 seniors at the time of the 1981 census reported that they had immigrated to Canada in the previous five years, and that as a result, 1.4 per cent of non-institutionalized seniors were recent immigrants to Canada. Furthermore, in 1981, 1.5 per cent of males and 1.6 per cent of females 65–74 years of age had immigrated to Canada in the previous five years. This phenomenon is even more pronounced in certain provinces and cities. In Ontario, the comparable figures were 1.9 per cent and 2.4 per cent while in British Columbia, the figures were 3.0 per cent and 2.1 per cent. In 1981, over 3 per cent of the seniors in Calgary, Toronto, and Vancouver had immigrated to Canada within the past five years. The research literature will be reviewed to examine these patterns of permanent elderly immigration to Canada.
In summary, in this article we will review progress in research on the geographic mobility of seniors over the past 25 years focusing on three issues: (a) trends in the rates of geographic mobility and resulting relocation and concentration of seniors in Canada, (b) trends in elderly seasonal migration, and (c) trends in the international migration of seniors to Canada from other countries. We will then assess the current status of research on elder migration in Canada. Finally, we will identify future challenges and unanswered research questions, and provide suggestions for future research.
Progress in Research on the Geographic Mobility of Seniors
Here, we will examine progress in research on the geographic mobility of seniors. First developments in theory and sources of data are examined followed by a focus on trends in relocation and concentration, seasonal migration, and international migration.
Developments in the Theorizing of Geographic Mobility
Walters (Reference Walters2002; see also Moore & Rosenberg, Reference Moore, Rosenberg, Marshall and McPherson1994) has suggested that although earlier models of later-life migration included push-pull, stress-threshold, place utility, and human capital/cost-benefit models, the life course model has since become increasingly common. The life course model suggests that patterns of geographic mobility vary across the life course, and that geographic mobility in later life is influenced by age-related life course events such as children leaving home, retirement, death of a spouse, declines in health, and reductions in income. Furthermore, life course trajectories at mid-life can influence moves in later life (Hayward, Reference Hayward2004).
Contemporary life course models focus on the relationship between age and the type of move undertaken (for example, amenity-driven moves when a person is younger-old and assistance moves or institutional moves when a person is older-old) while contemporary migration decision models tend to focus on the combination of place characteristics (i.e., the pushes and pulls of places of origin and destination) and the personal attributes of older movers and non-movers (Longino & Bradley, Reference Longino, Bradley, Binstock and George2006). Finally, the place identity model (Bradley & Longino, Reference Bradley, Longino and Uhlenberg2009) focuses on the meanings and emotional attachments people have for places and the consequences of these attachments for staying or moving. In short, contemporary theories of geographic mobility in older age emphasize demographic, economic, social, and psychological contingencies and dynamics that occur over a lifetime and that lead to patterns of moving and/or staying in place in older age.
Developments in Data Sources
In the mid-1980s, the primary data sources for studying the geographic mobility of Canadian seniors were the Canadian census, provincial health insurance registries, taxation records, and Old Age Security and Canada Pension Plan mailing lists. One of the advantages of the OAS and CPP mailing lists is that they can be used to identify moves outside of Canada. Since the mid-1980s, population surveys have matured and some of these surveys contain data on the geographic mobility of older Canadians. For example, studies of elderly mobility have been conducted by Sarma, Hawley, and Basu (Reference Sarma, Hawley and Basu2009) using the first six cycles (1994–1995 to 2004–2005) of the ongoing Canadian longitudinal National Population Health Survey; Sarma and Simpson (Reference Sarma and Simpson2007) drew on the longitudinal Aging in Manitoba study (with data from 1971 to 2001); Ostrovsky (Reference Ostrovsky2004) used the longitudinal Canadian Survey of Labour and Income Dynamics from 1996–2000; Hayward (Reference Hayward2004) made use of the Ontario Longitudinal Study on Aging 1959–1978; Gee, Kobayashi, and Prus (Reference Gee, Kobayashi and Prus2004) drew on the Canadian Community Health Survey of 2000–2001; Moore and Rosenberg (Reference Moore, Rosenberg, Marshall and McPherson1994, Reference Moore and Rosenberg1997) used both the 1991 Survey on Aging and Independence and the 1986 and 1991 Health and Activity Limitation Survey; and Bergob (Reference Bergob1995) employed the 1990 General Social Survey. The Canadian Longitudinal Study on Aging (CLSA) which, at the time of this writing, had begun the first wave of data collection, promises to produce a rich data set for future research (Martin-Matthews & Mealing, Reference Martin-Matthews and Mealing2009; Raina et al., Reference Raina, Wolfson, Kirkland, Griffith, Oremus and Patterson2009).
The Canadian Census from 1961 to 2006 was a major source of data on geographic mobility. The question: “Where did you live five years ago?” was first asked in the 1961 census (for details, see Northcott, Reference Northcott1988) and the question: “Where did you live one year ago?” was first asked in the 1991 census. The census data provided for analyses of movers and non-movers by demographic characteristics including age, gender, marital status, and education. The census data also provided for analyses by type of move, whether local, intra-provincial, interprovincial, or international in-migration and identified places of origin and destination within Canada. The federal government has discontinued the collection of census data on geographic mobility as of the 2011 census and instead plans to rely on surveys to obtain mobility data. This change will undermine the comparability of data from 2011 onward with the census databases from 1961 to 2006.
Trends in the Geographic Mobility and Concentration of Seniors
An extensive review of research from 1990 to 2000 on elder migration in the United States and Canada concluded that while seniors were less likely to move than non-seniors, elderly mobility had become increasingly common and tended to increase the concentration of seniors in selected locales (Walters, Reference Walters2002). Further, for later-life interstate migrants, the selection of popular destinations has expanded significantly, with the result that seniors are being dispersed somewhat more widely than in the past across an increasing range of amenity-rich destinations (Bradley & Longino, Reference Bradley, Longino and Uhlenberg2009). Nevertheless, the interstate migration of seniors in the United States tends to concentrate seniors in relatively few states (e.g., Florida) and in selected locales within states (e.g., Palm Beach in Florida or Phoenix in Arizona) (Bradley & Longino, Reference Bradley, Longino and Uhlenberg2009). This concentration of seniors is a result of both permanent and seasonal migration.
Although the relocation of seniors has produced striking concentrations of seniors in selected locations, many seniors do not move when they retire, and the motivations for aging in place (i.e., not moving) should be studied along with motivations for moving (Bradley & Longino, Reference Bradley, Longino and Uhlenberg2009). Further, Bradley and Longino (Reference Bradley, Longino and Uhlenberg2009) speculated about future patterns of later-life long-distance moves and concluded that it is not clear that the baby boomer cohort will continue the existing pattern of later-life amenity-driven interstate retirement moves. Although some boomers will be better equipped to make amenity moves at retirement than previous cohorts, not all will be well-off, and many may delay retirement while others may be content to age in place in locations where they have long resided and which have become increasingly senior-friendly.
Most Canadian seniors “age in place” and choose not to move over long periods of their older years. An examination of 2006 census data on geographic mobility shows that over 70 per cent of the older population did not move in the previous five years, and over 90 per cent did not move in the previous year. For the minority of seniors who did change their residence, a little over half of all moves reported by older persons were local moves within the same city, town, village, township, or municipality (Northcott & Petruik, Reference Northcott and Petruik2010). The following study illustrates this common pattern. Everitt and Gfellner (Reference Everitt and Gfellner1996) studied elderly voluntary movers and stayers in Southwest Manitoba and found that most seniors prefer to age in place in their homes or age “near” place by moving, for example, from the farm to small nearby communities. They note that personal identity and place identity tend to be fused for older persons, and as a result, most seniors prefer to stay within their local communities either as non-movers or as local short-distance movers.
Many seniors have lived in the same place for years. Further, they often own their homes outright having previously paid their mortgage in full. While home equity might provide seniors with the economic means to relocate, owning one’s home is also a powerful motivation not to move. To this point, Ostrovsky (Reference Ostrovsky2004) examined the longitudinal Canadian Survey of Labour and Income Dynamics from 1996–2000 to determine whether the residential mobility of older Canadians is affected by a life-cycle-related motivation to spend the equity they have previously accumulated in their homes. The findings of this study were consistent with studies in the United States and Europe that have found that seniors generally are not motivated to move out of their homes to free up their accumulated housing equity for general spending. Seniors may move for other reasons: for example, they may be forced to move by poor health. This tendency to view housing equity as a non-spendable asset might explain in part the tendency of older persons to age in place.
The concentration of seniors in a given locale is a function of seniors’ mobility patterns, both staying and moving, and is also a function of the movement patterns of non-seniors. Moore and Pacey (Reference Moore and Pacey2004; see also Bergob, Reference Bergob1995; Moore & Rosenberg, Reference Moore, Rosenberg, Marshall and McPherson1994; Moore, Rosenberg, & McGuinness, Reference Moore, Rosenberg and McGuinness1997) examined the relative impact of aging in place and net migration on population aging for Canadian provinces and cities from 1991–1996 and 1996–2001. Although the aging of the population is primarily a function of the resident population growing older without moving, net migration can have significant effects on the proportion of the population that is seniors. In particular, the net migration of the more mobile younger population is an important factor in population aging. Outmigration of non-seniors increases population aging at the origin and decreases population aging at the destination. Given that the migration of non-seniors is highly affected by economic conditions (i.e., a weak local economy results in net outmigration while a strong local economy results in net in-migration), it follows that seniors tend to become concentrated in areas where the local economy is weaker as younger people leave to find jobs elsewhere. Of course, in time, young migrants age and become seniors, and patterns of migration earlier in life can have consequences in older age. For example, Liaw and Qi (Reference Liaw and Qi2004) showed that persons who migrated from one province to another at some time in their lives and who are now seniors have higher incomes than persons who have stayed in their province of birth.
It is well-known that seniors who move tend to concentrate in selected locales such as Victoria, St. Catharines, or the Okanagan Valley (Moore, Rosenberg, & McGuinness, Reference Moore, Rosenberg and McGuinness1997; Northcott, Reference Northcott1988, Reference Northcott1984; Northcott & Milliken, Reference Northcott and Milliken1998). What is not as well-known are the social and economic implications of these concentrations of older persons. Joseph and Cloutier (Reference Joseph and Cloutier1991; see also Dahms, Reference Dahms1996) examined the (net) in-migration of elderly persons into rural Grey County, Ontario, and their concentration in villages and towns in the county. They concluded that there is a need for analysis of migration impacts on demand for services at the local level. Hodge (Reference Hodge1991) noted that retirees tended to be an economic asset in smaller communities in British Columbia, and Li and MacLean (Reference Li and MacLean1989) arrived at the same conclusion in their study of small towns in Saskatchewan.
Serow (Reference Serow2003) made a similar point in his review of studies examining the economic consequences of retiree concentrations in the United States and Canada; however, Serow noted that while the economic impact of the in-migration of amenity-seeking retirees tends to be positive in the short term, the long-term implications are not clear. He suggested that longitudinal studies of the impact of retiree migration to specific destinations are needed with samples of sufficient size to study impacts at the local level. Serow declared that “it is at the municipal and county level where economic changes will be most clearly felt” (p. 901). Similarly, Moore and Pacey (Reference Moore and Pacey2004; see also Moore & Rosenberg, Reference Moore, Rosenberg, Marshall and McPherson1994; Moore, Rosenberg, & McGuinness, Reference Moore, Rosenberg and McGuinness1997) noted that the effect of net migration for population aging tends to be of greater consequence for smaller and less populated areas. Nevertheless, Moore and Pacey (Reference Moore and Pacey2004) wrote:
From a public policy viewpoint, the changing geography of aging should be viewed as a national, as much as a regional or local issue. Given the significant service demands of seniors, particularly in the areas of health care and social services, the changes in population aging generated by shifts in economic performance have widespread implications. … Without serious attempts by central governments to address disparities among areas, the risk is enhanced that the landscape of have and have-not communities will become even more pronounced. (p. S19; italics in the original)
Finally, Joseph and Martin-Matthews (Reference Joseph, Martin-Matthews, Marshall and McPherson1994, p. 23; see also Hodge, Reference Hodge1991, Reference Hodge2008) observed that communities will vary substantially depending on whether the increase in the concentration of seniors is primarily a result of out-migration of younger people, aging in place, or in-migration of older people. Communities will also vary depending on whether they have attracted older migrants who have lived near by and who come seeking better housing and services or younger migrants coming from longer distances seeking specific amenities.
Although the number of seniors who move from one province to another in any given year is relatively modest, and although an examination of census data from 1961 to 2006 showed that seniors are not increasingly likely to change their province of residence (Northcott & Petruik, Reference Northcott and Petruik2010), nevertheless, interprovincial migration can have substantial cumulative consequences for the redistribution and concentration of seniors in Canada. The outmigration of seniors from a province has the potential to reduce the concentration of seniors in that province unless there is also a counterbalancing outmigration of non-seniors. Similarly, the in-migration of seniors to a province has the potential to increase the concentration of seniors in that province unless there is also a counterbalancing in-migration of non-seniors. If non-seniors leave a province at a greater rate than seniors do, then the seniors remaining behind will represent an increasing concentration of that provincial population. In Canada, senior and non-senior migration streams tend to be similar (that is, seniors and non-seniors tend to leave the same places and go to the same destinations) with the result that these streams tend to be counterbalancing to a degree at least with respect to impact on provincial age structures (see, for example, Northcott & Milliken, Reference Northcott and Milliken1998).
Researchers have found, too, that persons who have moved earlier in their lives are more likely to move in older age (Hayward, Reference Hayward2004). Similarly, Liaw and Ledent (Reference Liaw and Ledent1988; see also Newbold & Liaw, Reference Newbold and Liaw1990) observed that seniors living in the province of their birth were less likely to move to another province than seniors who have already at some time in their lives moved to another province. Moreover, seniors who migrated to Canada at some time in their lives were more likely to move interprovincially.
In the period 1976–1981, senior interprovincial migrants were almost two and a half times as likely to move west as they were to move east, with Ontario and especially British Columbia showing net gains of seniors while all other provinces showed net losses (Northcott, Reference Northcott1988). Similarly, Liaw and Ledent (Reference Liaw and Ledent1988) noted that seniors who move interprovincially tend to move westward, with Atlantic seniors concentrating in Ontario and seniors from Ontario and the Prairies concentrating in British Columbia. Newbold and Liaw (Reference Newbold and Liaw1990; see also Liaw & Ledent, Reference Liaw and Ledent1988) commented on the “general westward net drift of migrants in Canada” (p. 24) and characterized British Columbia and the Pacific Coast more generally, as the “end of the line” in the sense that people who migrate there tend to stay and people who leave there tend to return.
Geographic mobility is very much a function of age or position in the life course. While many elderly persons do not move, Litwak and Longino (Reference Litwak and Longino1987; see also Serow, Reference Serow1987; Wiseman, Reference Wiseman1980) discussed three types of life-course-related moves undertaken by seniors who do move. The first type of move is related to retirement in younger old age and typically involves a move to an amenity-rich destination. The second type of move tends to come later in old age and is driven by declines in health and an increased need for support typically offered by family members. The third type of move involves institutionalization. Moore and Rosenberg (Reference Moore and Rosenberg1997) showed that about one third of moves undertaken by older seniors fit Litwak’s and Longino’s second type of move, that is, a move designed to access greater social support. It is important to note that not all non-movers among seniors are voluntary stayers – some may prefer to move but lack the resources to do so and can be considered to be “blocked movers” (Moore & Rosenberg, Reference Moore and Rosenberg1997).
When older seniors move, a common motivation, as we have noted, is to access support from family members. Firbank and Johnson-Lafleur (Reference Firbank and Johnson-Lafleur2007) examined the process by which elderly persons in the province of Quebec who lived independently at home came to live with a caregiver, usually an adult child (a living arrangement referred to as intergenerational cohabitation). The authors pointed out that this is a fairly common response to a senior’s declining health and increasing dependency. Further to this point, Sarma, Hawley, and Basu (Reference Sarma, Hawley and Basu2009; see also Trottier, Martel, Houle, Berthelot, & Légaré, Reference Trottier, Martel, Houle, Berthelot and Légaré2000) examined transitions from independent to intergenerational and institutional living arrangements for Canadian seniors using data from the first six cycles (1994–1995 – 2004–2005) of the Canadian longitudinal National Population Health Survey. They showed that publicly funded formal home care services provided to seniors living independently reduced the likelihood of moving into an institution but had no effect on the likelihood of transition to an intergenerational living arrangement. In addition, they showed that the determinants of a transition from independent living to intergenerational or institutional living included a lack of social support and informal care, not being currently married, poor health, low income, and older age. For a similar conclusion, see Sarma and Simpson (Reference Sarma and Simpson2007; see also Tomiak, Berthelot, Guimond, & Mustard, Reference Tomiak, Berthelot, Guimond and Mustard2000), who analyzed data (from 1971–1996) from the longitudinal Aging in Manitoba study.
As previously stated, seniors who move tend to undertake different kinds of moves depending on where they are in their life course. Retirement migration triggered by withdrawal from the labour force typically involves the younger elderly and may take the following forms: return migration to the place where one was born and raised, kinship migration to locate near adult children and grandchildren, or amenity migration involving a move to a place deemed to offer a more attractive quality of life. Later in the life course, assistance migration may be triggered by an increased need for informal social supports (e.g., assistance-motivated kinship migration) and/or formal supports (e.g., professional health care and/or institutional living arrangements) and might involve return migration following a previous amenity migration. Although many seniors will choose to stay and not move, it is noteworthy that some seniors may desire to move but be unable to do so because of unfavourable circumstances (that is, blocked migration). Finally, towards the end of life and typically in the older old years, forced migration might involve an involuntary move to a nursing home, for example.
While the aforementioned life-course-related sequence of moves tends to focus on relatively permanent relocation, seniors – and especially younger and healthier seniors – often make moves of a temporary nature, which we discuss next.
Trends in Seasonal Migration of Seniors
Many elderly Canadians spend up to six months each winter in the Sunbelt of the United States (for example, in Florida, Arizona, California, or Texas). Returning to Canada within six months maintains their eligibility for provincial health care. They typically have a home in Canada and spend their time in sunnier climes in U.S. accommodations they own or rent, or they live in recreational vehicles (RVs) they drive to U.S. destinations. Accommodations include homes, condominiums, rented apartments, “park models” in trailer parks, and RVs (motor homes, trailers) driven or pulled to RV parks or campgrounds. Some have sold the family home in Canada but kept the cottage at the lake and have become permanent vacationers, spending at least six months in the cottage in Canada and up to six months in the American Sunbelt (Marshall, Longino, Tucker, & Mullins, Reference Marshall, Longino, Tucker and Mullins1989). While most snowbirds are relatively sedentary and spend the winter in a specific location, Counts and Counts (Reference Counts and Counts2001) observed that some seniors are nomadic “full-timers” who live in their recreational vehicles permanently and have no other home, although they typically maintain a Canadian address in order to qualify for provincial health care.
Considerable work was done in the 1980s to examine concerns that Canadian (and American) snowbirds would overwhelm local health care and social services to the detriment of local populations in locations where snowbirds tended to flock. These studies showed that the Canadians tended to be positively selected, that is, they tended to be the younger, healthier, wealthier, and married seniors who made few demands on health and social services while in the United States, preferring to return to Canada for health care, and typically gave up their travels when their health deteriorated (Daciuk & Marshall, Reference Daciuk and Marshall1990; Longino & Marshall, Reference Longino and Marshall1990; Marshall et al., Reference Marshall, Longino, Tucker and Mullins1989; Tucker, et al., Reference Tucker, Mullins, Béland, Longino and Marshall1992).
Canadian snowbirds in the United States number in the tens of thousands each winter during the months of November to April. Indeed, the number may be in the hundreds of thousands – exact figures do not currently exist, and the numbers tend to rise and fall depending in part on the exchange value of the Canadian dollar. Mexico may be becoming a major destination for Canadians seeking a cheaper alternative to the U.S. Sunbelt (Coates, Healy, & Morrison, Reference Coates, Healy and Morrison2002). Francophone seniors from Quebec tend to flock to the Atlantic Coast of Florida while Anglophone seniors from Ontario tend to congregate on the Gulf Coast of Florida (Longino, Marshall, Mullins, & Tucker, Reference Longino, Marshall, Mullins and Tucker1991; Tucker et al., Reference Tucker, Mullins, Béland, Longino and Marshall1992). Seniors from the Prairie Provinces tend to flock to Arizona, California, and Texas. These seniors spend Canadian dollars in the United States during the winter months, rather than in Canada, benefitting the local economies where they congregate with some cost to the local economies back home (Tucker et al., Reference Tucker, Mullins, Béland, Longino and Marshall1992). Furthermore, when the snowbirds have problems, they tend to return to Canada. The result is that the “burden” of supporting seniors who need public support remains in Canada despite the considerable time that seniors spend in the United States along with the accompanying outflow of funds from Canada. Furthermore, Joseph and Martin-Matthews (Reference Joseph, Martin-Matthews, Marshall and McPherson1994) noted that the southward flight of elderly snowbirds each winter diminishes the volunteer pool especially in rural places in Canada and reduces the assistance that would otherwise be available to aging seniors in need who remain in Canada during the winter months.
There is still much to be learned about the snowbird phenomenon. Coates and colleagues (2002) observed that:
This inchoate, uncoordinated, and little-studied aspect of North American life may well prove to be one of the most important social and economic developments of the new millennium, challenging national loyalties, enriching or disrupting host and sending communities, forming an expanded continental consciousness, and spreading the influence of continental integration across North America. (p. 448)
Apparently, the seasonal migration of elderly Canadian snowbirds to southern destinations remains a greatly underestimated and understudied phenomenon.
Trends in International Migration of Seniors
There are different types of international later-life movers including return migrants to their country of birth following engagement in the labour force of the host country; family-joining (kinship) migrants – older persons moving across borders to live with or live near, and/or to seek assistance from, family members (e.g., elderly immigrants who come to Canada to join their adult children who have immigrated previously); amenity-seeking migrants (e.g., Canadian senior snowbirds in Florida or Arizona); and cyclical international migrants (e.g., persons who have immigrated to Canada earlier in their life and who in older age return to their home country semi-permanently or seasonally) (Longino & Bradley, Reference Longino, Bradley, Binstock and George2006; Warnes, Reference Warnes and Uhlenberg2009).
The amenity-seeking migration of retirees has a long history originating in the early twentieth century in Europe and the United States (and for the European elite, going back centuries further in time) (Longino, Reference Longino and Birren2007; Longino & Warnes, Reference Longino, Warnes and Johnson2005). Longino and Warnes (Reference Longino, Warnes and Johnson2005) have suggested that retirement migration is becoming increasingly common and that current retirement destinations are becoming more widely dispersed. Not only are retirees moving to amenity locales within their countries of residence, but also, they are increasingly moving to amenity areas in other countries (e.g., British retirees in Spain, Canadian retirees in the American Sunbelt, American retirees in Mexico). Longino and Warnes (Reference Longino, Warnes and Johnson2005) concluded that “the global dispersal of [retirement] destinations is … well underway” (p. 544) including such “exotic” locations as the South Pacific and India.
Warnes (Reference Warnes and Uhlenberg2009) noted the rise of “transnationalism” wherein seniors reside in more than one country – for example, a Canadian senior might have homes in both Canada and the United States, an American senior might have homes in both the United States and Mexico, a British senior might have homes in both the United Kingdom and Spain, and a person who migrated earlier in life might in their older age maintain residences in both their country of origin and their adopted country. International retirement migration can be substantial and consequential. For example, Warnes (Reference Warnes and Uhlenberg2009) explained that in 2005, 8.6 per cent of U.K. state pensions were paid to oversees addresses covering a global range of countries, although most overseas pensioners from the United Kingdom resided in Australia, Canada, the United States, and Ireland. The transfer of state benefits to older persons living outside the United Kingdom amounted to over two billion pounds in 2005. It is expected that the percentage of British pensioners living outside of the United Kingdom will increase over the foreseeable future and that this phenomenon will become increasingly widespread involving seniors in many different countries.
Kobayashi and Preston (Reference Kobayashi and Preston2007) remarked that transnationalism involving families with homes in two different countries with different patterns of family dispersal across the two homes depending on position in the life course is a growing phenomenon throughout the Asia-Pacific region. For example, a father who is a recent immigrant to Canada from Hong Kong may return to Hong Kong to work while the mother, also an immigrant to Canada, raises the couple’s children in Canada where they receive their education. Later, some of the children may go to Hong Kong to work while their parents retire in Canada.
Turning to the migration of seniors to Canada from other countries, the 2006 Census of Canada indicated that 10,305 seniors moved to Canada within the previous year, and 38,175 seniors moved to Canada within the past five years from other countries (compared to 10,073 in the years 1956–1961). Almost one in every 100 seniors had migrated to Canada in the previous five years, and one in every 400 had migrated in the previous year (Northcott & Petruik, Reference Northcott and Petruik2010). This census also showed that 28 per cent of the population aged 65 and older had been born outside of Canada. Almost two thirds (66%) of these immigrant seniors had lived in Canada for more than 35 years while six per cent had come to Canada within the past 10 years (Statistics Canada, 2010). In any given year from 1995 to 2004, two to four per cent of immigrants arriving in Canada were aged 65 or older (Turcotte & Schellenberg, Reference Turcotte and Schellenberg2007, p. 23). The countries of origin of older immigrants to Canada have changed in recent decades with an increasing percentage coming from Asia. In 2001, 19.1 per cent of immigrant seniors in Canada had been born in Asia, up from 5.6 per cent in 1981 (Turcotte & Schellenberg, Reference Turcotte and Schellenberg2007, p. 23).
In the United States, Treas and Batalova (Reference Treas, Batalova and Uhlenberg2009) observed that of foreign-born seniors, most had come to the United States years ago when they were much younger, as is the case in Canada, while one in eight had immigrated to the United States in the past 10 years, many of these coming to join adult children who had migrated previously. Treas and Batalova reported that 4.4 per cent of immigrants coming to the United States in 2005 were elderly (compared to 2.3 per cent for Canada in 2004). Given current migration patterns, the proportion of seniors in the United States who are foreign-born and the diversity of these seniors in terms of country of origin are expected to increase well into the twenty-first century. Treas and Batalova suggested that persons who immigrate when older tend to remain economically dependent on their families and have more difficulty adapting to their host society than persons who immigrate when they are younger.
Most elderly immigrants coming to Canada settle in Ontario, British Columbia, and Alberta (Northcott & Milliken, Reference Northcott and Milliken1998). In the early 1990s in British Columbia, far more elders migrated to the province from other countries than migrated from British Columbia to other countries (Northcott & Milliken, Reference Northcott and Milliken1998). Little is known about how many Canadian seniors have moved out of Canada and settled in various other countries although a literature on Canadian retirees in other countries is beginning to emerge (see, for example, Banks, Reference Banks2004, and Truly, Reference Truly2002, for studies of Canadian and American retirees living permanently in the Lake Chapala Riviera in the State of Jalisco, Mexico).
In an effort to determine if immigrant seniors who have come recently to Canada exhibit the “healthy immigrant effect” observed in younger migrants to Canada, Gee and colleagues (2004) examined the Canadian Community Health Survey of 2000–2001. They noted that older recent immigrants tend to have worse health than persons born in Canada, in contrast to younger migrants who tend to have better health. This differential between older recent immigrants and Canadian-born seniors was largely explained by socio-demographic factors (e.g., lack of English-language proficiency) as well as socioeconomic (e.g., low income) and health-related lifestyle (e.g., smoking) factors.
In summary, there has been a growing body of literature on the international migration of seniors including amenity-seeking elderly immigrants, rising trends of elderly persons living transnationally, and preferred destinations of elderly immigrants to Canada.
Current Status of Research on the Geographic Mobility of Seniors in Canada
In the mid-1980s, there was a need for better theoretical, conceptual, and empirical explanations of seniors’ geographic mobility. Since that time, there has been more emphasis placed on conceptualizing and examining the different types of moves that seniors undertake, including non-movers who stay and “age in place”. Furthermore, there has been more emphasis placed on examining the determinants of the various types of moves and their consequences. Nevertheless, research emphasis has focused more on movers than non-movers and on the determinants rather than the consequences of moves. More attention needs to be placed on theorizing, conceptualizing, and researching the different types of non-movers and the determinants and consequences of not moving for the majority of seniors who age in place. In addition, more research should consider the consequences of moving for seniors who do move.
More attention must also be paid to different levels of analysis including the individual senior, the senior as a part of a couple relationship and/or a family network, the communities in which seniors stay or to which they move, and the province and nation which is shaped in part by the dynamics of age-specific population relocation and concentration. Most work done to date has taken the individual senior as the unit of analysis. This has been the case for the geographic mobility data collected in the Canadian censuses from 1961 to 2006 and is the usual case when social surveys are conducted. Given that a senior who is married typically moves not as an individual but as a couple, there is a need for couple-level theorizing and data. Furthermore, given that single seniors often make a move that takes other family members into account, there is a need for theorizing and data collection that explore the dynamics of moving or staying in the context of the family network. Although patterns of movement are fairly well understood, decisional dynamics leading to moving or staying – and the experiential consequences of moving or staying for seniors and seniors’ families – are not as well understood. Finally, while some work has been done examining the consequences of senior mobility for sending and receiving communities as a whole, more needs to be done exploring the consequences of the relocation and concentration of seniors for communities, provinces, regions, and the nation as a whole.
Increasingly, researchers are applying a life course perspective to theorize and study the sequence of moves undertaken at various junctures in early adulthood, middle age, and old age, along with their determinants and consequences. Given that a 20-year-old individual’s move and a 70-year-old individual’s move to British Columbia from the farm in Saskatchewan or to return home to Nova Scotia from Ontario are different types of moves with different determinants and consequences, the life course perspective seems to be quite useful. Of course, it is important to develop a range of theoretical and conceptual perspectives.
In the mid-1980s, there was a need for social surveys to obtain more insight into the personal factors related to moving or staying. Although the census had provided considerable data and insight, census data were limited to demographic analyses of geographic mobility including frequency of moves, distance of moves (whether local, longer distance within a province or from one province to another, and international in-migration), origins and destinations and resulting net migration, and the demographic correlates of movers and non-movers (including age, gender, marital status, level of education, etc.). Social surveys and in-depth interviews were needed to explore the social psychological aspects of geographic mobility including the motivational and decisional dynamics of moving or staying and the resulting experiential consequences. Since that time, a number of major surveys have been conducted, and many of these have been made available to researchers through Research Data Centres established on a number of university campuses across Canada.
These surveys will become increasingly important given the plan to eliminate census questions on geographic mobility from the 2011 and subsequent censuses. The implementation of longitudinal surveys such as the National Population Health Survey and the forthcoming Canadian Longitudinal Study on Aging are important vehicles for examining the sequence of moves (or of not moving) in old age and their determinants and consequences. Nevertheless, in these surveys, the unit of analysis also is typically the individual, and we still need to know more about the familial dynamic involved in decisions to stay or to move in old age. Furthermore, surveys such as the National Population Health Survey are not centrally dedicated to the study of geographic mobility and so provide only limited data on elder mobility.
There is a need for a greater variety of methods to be applied to the study of elder migration. To date, there has been a heavy reliance on census and survey data and, as we have noted, on the individual as the unit of analysis. Qualitative studies have been rare and would provide considerable insight into the dynamics of decision making regarding moving or staying and the experiential consequences of those decisions. Relevant qualitative methods include in-depth interviews, ethnographies, narratives, and biographies. Ethnographies and case studies also have the advantage of facilitating studies of aggregate units of analysis such as communities. For example, a comparative study of the senior-rich Okanagan and a senior-deficient resource community such as Fort McMurray in northeastern Alberta might shed light on the consequences at the community level of concentrations of seniors in comparison with a relative absence of seniors. To date, there have been relatively few regional studies done in Canada, and indeed, there may be an academic preference for national and international studies and a subtle bias against more “parochial” regional and local studies. In short, while much has been accomplished, much remains to be done.
Challenges, Unanswered Questions, and Suggestions for Future Research
With the demise of the mobility questions asked on censuses in Canada from 1961–2006, perhaps the greatest challenge for future researchers will be to obtain adequate data on the demographics of geographic mobility in Canada. The census provided comprehensive data on the movement of the Canadian population by age group within provinces, between provinces, and between cities. The censuses also counted recent arrivals to Canada. It is unlikely that any survey will be designed in the future to provide as comprehensive a picture of the demography and geography of the movement and stability of the Canadian population. Nevertheless, surveys do have an advantage over the census when it comes to obtaining data of a more social psychological nature.
The challenge is to find the funding to design and implement surveys dedicated to exploring the social psychology of moving and staying; however, large-scale structured surveys administered over the telephone, for example, with fixed questions and a limited choice of answers have their limitations, and response rates to surveys of this nature have been declining. The difficulty now is to find ways to increase participation rates in social surveys and also to apply relevant qualitative methodologies to investigate the social psychological aspects of moving and staying. Surveys are not equally well-adapted to studying all sub-populations. For example, institutionalized seniors, seniors who are frail, or who have hearing difficulties or suffer from dementia, are often excluded from surveys. The challenge is to find appropriate and effective ways to study forced moves at the end of life involving frail seniors, family caregivers, hospitals, nursing homes, and assisted living facilities. Another challenge relates to the need for aggregate-level analyses of communities and provinces impacted by age-specific geographic mobility. Finally, it is important to note that some unanswered questions may be difficult or impossible to answer, given current data limitations.
With respect to the geographic relocation and concentration of seniors in Canada, we need to know more about the sequence and dynamics of moves across the older years. We need to know more about blocked moves and involuntary stayers – not all who age-in-place do so by choice as is often assumed. We need to conduct studies of senior mobility at different levels of analysis including the older individual, older couple, extended family, and community. Assistance migration needs to be better understood – how common, for example, is moving to be near kin for assistance or to live with kin for assistance, and what are the family dynamics involved in assistance moves? It has been suggested that amenity destinations for retirement migrants may be becoming more dispersed. Is this so in Canada, and how wide or selective is that dispersal? Will the boomer cohort (born 1946–1966) retire later and move less? Will they disperse more widely both within Canada and outside of Canada? To what extent will they turn into snowbirds? Will Canada’s cities and towns become more “senior friendly” as they adapt to population aging, and will amenity migration decline and aging in place increase as a consequence? That is, will the development of “senior friendly” communities reduce pressures to move and increase motivations to stay?
What are the economic and social consequences and implications of senior and non-senior migration patterns and resulting concentrations of seniors? We need to know more about the economies of aging in local contexts such as rural areas and small towns as well as larger cities and provinces as a whole. We need to know more about the consequences of senior migration for the distribution and mobilization of human capital – for example, the volunteer labour pool. That is, we need to address the following questions: What are the economic and social costs and benefits of outmigration of seniors and non-seniors for local communities? What are the costs and benefits of in-migration of seniors and non-seniors for local communities? What are the costs and benefits of in-migration of seniors who are amenity migrants, return migrants, assistance migrants, and what are the short-term and long-term costs and benefits?
We must study these impacts over time. To this point, Serow (Reference Serow2003) suggested that longitudinal studies of the impact of retiree migration to specific destinations are needed with samples of sufficient size to study impacts at the local level. Serow (Reference Serow2003) also asked about the kind of jobs seniors create, suggesting that they may be low-skill low-wage jobs. If seniors create jobs, exactly what jobs do they create and what jobs are lost? What are the local impacts of concentrations of seniors for personal spending, tax revenues, and tax expenditures? Finally, what are the implications of an “inequitable” distribution of seniors across Canada’s social landscape? Can and should a more equitable distribution of seniors be achieved?
Turning to the phenomenon of seasonal migration, we need more information on the number and types of snowbirds, and their motivations for starting and stopping their seasonal migration What are the economic and social costs and benefits of annually traveling for six or so months at a time for individual seniors, their families, their communities of origin in Canada, and their host communities in the US and elsewhere? Coates and colleagues (2002) suggested that we need to study the implications of seasonal migration for several reasons: economically, politically as a challenge to national loyalties, and socially as a force for continental integration and an expanded continental consciousness.
Finally, turning to the international migration of seniors, we need to know more about the social and economic implications of immigrants moving to and from Canada in their older years. Are Canadian seniors exporting themselves to nicer and cheaper locations outside of Canada and with what implications? What ties are broken and maintained? How many Canadian seniors continue to receive their Old Age Security, Canada Pension Plan, and other pension income in Canada (and so appear to be Canadian residents) but live outside of Canada and spend those dollars outside of Canada? Can seniors be enticed to stay in Canada and spend in Canada? Are non-Canadian seniors seeking to retire in Canada because of its generous health care programs and high quality of life? With respect to older immigrants who have moved recently to Canada, we need to know more about the unhealthy older immigrant effect (Gee et al., Reference Gee, Kobayashi and Prus2004) that contrasts the healthy immigrant effect observed in younger immigrants coming to Canada.
We also need to know more about the economics of transnationalism: that is, of seniors who are citizens of two or more countries and who may live in different countries at different times of their lives. Are transnational Canadians working in other countries but retiring in Canada? Will seniors who migrate to Canada maintain residences in Canada and their countries of origin and with what consequences for spending and provision of services? Will seniors emphasize their country of origin when they are healthy, and their adopted country of Canada with its generous health care system when not healthy? Will pension dollars be spent in the countries of origin and tax dollars consumed in Canada? To what extent is there a parallel to transnationalism within Canada – that is, a transprovincialism where seniors maintain residences in two provinces within Canada?
In short, although we have learned much about the geographic mobility of elderly Canadians in the last quarter century, a number of enticing questions remain for researchers to explore as the boomers enter old age over the next two decades.
